Here you can find out how to score the items in NOSACQ50.
How to score the items in NOSACQ50
Option 1
Submit your raw data to the NOSACQ50 development team, and you will receive a short report with your results. See section regarding submitting your data. If you submit your data to the NOSACQ development team it will be possible to receive a short summary report where your data is compared to the international database.
Option 2
You can calculate the results yourself by following the procedures below, as well as using SAS or SPSS
The questionnaire is organized in 7 dimensions with between 69 items each. The items can be divided into two groups – those asked in a positive way and those items asked in a reversed (negated) way.

Positively formulated items

Reversed formulated items

Dimension 1 management safety priority and ability (9 items):

A1, A2, A4, A6, A7

A3, A5, A8, A9

Dimension 2 – management safety empowerment (7 items):

A10, A11, A12, A14, A16

A13, A15

Dimension 3 – management safety justice (6 items):

A17, A19, A20, A22

A18, A21

Dimension 4 – workers’ safety commitment (6 items):

A23, A24, A27

A25, A26, A28

Dimension 5  workers’ safety priority and risk nonacceptance (7 items):

A33

A29, A30, A31, A32, A34, A35

Dimension 6 – Peer safety communication learning, and trust in safety ability (8 items):

A36, A37, A38, A39, A40, A42, A43

A41

Dimension 7 – workers’ trust in efficacy of safety systems (7 items):

A44, A46, A48, A50

A45, A47, A49

All the items in the questionnaire are answered in the same way (Lichert scale 14), but have to be scored dependent of the formulation of the question:

Strongly disagree

Disagree

Agree

Strongly agree

Score for positive items

1

2

3

4

Score for reversed items

4

3

2

1

In order to calculate the mean score for a single dimension the easiest way is to enter the observed values into a spreadsheet like Excel and then subsequent calculate the true means.
For dimension 1 the true mean for one person will be:
(A1+A2+(5A3)+A4+(5A5)+A6+A7+(5A8)+(5A9)) / Number of answered items
Only answered items will be entered into the calculation. If a person has answered all 9 items the denominator ‘Number of answered items’ for dimension 1 will be 9. If a person has only answered 7 items the denominator ‘Number of answered items’ will be 7. (See example below).
However, if a person has answered less than 50% of the items in a dimension, all his / her answers in that dimension should be ignored and not be included in the total mean. (See example below).
When all the participants' data have been entered into the spreadsheet and their means for the dimensions have been calculated, a total mean for the population has to be calculated for each of the 7 dimensions as a 'mean of the mean'.
Example – how to calculate the 'mean of the mean' for dimension 1:
5 persons have the following scores on dimension 1: 2.67; 2.33; 2.44; 2.56; 2.67
The total mean for that population will then be (2.67+2.33+2.44+2.56+2.67)/5 = 2.53
DIMENSION 1 – EXAMPLE

QUESTION NUMBER
r = This data has been reversed

PERSON NUMBER

A1

A2

A3r

A4

A5r

A6

A7

A8r

A9r

MEAN (per person)

1

3

4

4

4

3

3


3

4

28/8 =3.50

2

3

4

2

3

4

3

2

2

1

24/9 =2.67

3

1

2

1


4





No mean calculated
More than 50% missing

Total mean score for this population of three persons and dimension 1

3.50+2.67 / 2 = 3.08

Proceed with same procedure for each of the remaining dimensions.
When calculating the means for each person you can use the following calculations:
Dimension 1: (A1+A2+(5A3)+A4+(5A5)+A6+A7+(5A8)+(5A9)) / Number of answered items
Dimension 2: (A10+A11+A12+(5A13)+A14+(5A15)+A16) / Number of answered items
Dimension 3: (A17+(5A18)+A19+A20+(5A21)+A22) / Number of answered items
Dimension 4: (A23+A24+(5A25)+(5A26)+A27+(5A28)) / Number of answered items
Dimension 5: ((5A29)+ (5A30)+ (5A31)+ (5A32)+A33+(5A34)+ (5A35)) / Number of answered items
Dimension 6: (A36+A37+A38+A39+A40+(5A41)+A42+A43) / Number of answered items
Dimension 7: (A44+(5A45)+A46+(5A47)+A48+(5A49)+A50) / Number of answered items
Reliability
Testing the reliability of the results within a scale should be done seperately for the worker and leader responses. We have used Cronbach's Alpha as an indicator for the reliability of the results. A Cronbach's Alpha below 0,67 is considered poor; 0.67  0.8 is considered acceptable; and above 0.8 is good.
In SPSS you can run the test under: Analyze  Scale  Reliability analysis
RELIABILITY
/VARIABLES=a1 a2 a3 a4 a5 a6 a7 a8 a9
/SCALE('Dim 1  Management safety priority and ability') ALL
/MODEL=ALPHA
/SUMMARY=MEANS
/SUMMARY=TOTAL.
RELIABILITY
/VARIABLES=a10 a11 a12 a13 a14 a15 a16
/SCALE('Dim2  Management safety empowerment') ALL
/MODEL=ALPHA
/SUMMARY=MEANS
/SUMMARY=TOTAL.
Etc. etc. for each dimension.